Work in this group focuses on the elucidation of the developmental signals that are required to regulate cell proliferation. To this end, we have been studying a gene, roughex(rux), that is required to arrest cells in G1 in the developing compound eye of Drosophila. Genetic and biochemical interactions suggest that rux is acting as a novel type of cyclin kinase inhibitor, a class of proteins recently identified in both yeast and mammalian systems. We have recently shown that overexpression of Rux in vivo inhibits mitosis and leads to endoreduplication. Mutations in rux interact genetically with the G2 Cyclin A, in that loss-of-function mutations in cycA act as dominant suppressers of the rux phenotype. In addition, overexpression of Cyclin A phenocopies rux, and co-expression of Rux reverses this. Finally, overexpression of Rux blocks mitosis by inhibiting accumulation of the G2 Cyclins A and B. Rux itself is expressed in a dynamic pattern in the developing eye. Expression begins in a band anterior to the morphogenetic furrow corresponding to the G1 domain that is absent in the mutant. Expression is decreased in the furrow, and then Rux protein is seen to reaccumulate behind the furrow and is rapidly down-regulated in cells that re-enter S phase in this region. Down-regulation of Rux protein is mediated by Cyclin E expression in these cells; ectopic overexpression of Cyclin E in cells constitutively expressing Rux behind the furrow leads to destabilization of Rux protein and accumulation of cyclins A and B. We propose a model whereby Rux is required to inhibit Cyclin A activity in G1, possibly by targeting Cyclin A for destruction. As cells enter S phase, inactivated by a Cyclin E-dependent mechanism, releasing cyclin A for its mitotic role.